Presented at the 14th International Conference of the Association of Institutes for Tropical

Veterinary Medicine (AITVM)

Johannesburg, South Africa

25-29 August 2013

F Shija, G Misinzo, H Nonga, LR Kurwijila, K Roesel and D Grace

Food-borne diseases are a threat and are

responsible for 33-90% cases of mortality to

children

Bacterial milk contamination causes:

◦ Milk spoilage

◦ Milk-born zoonotic diseases

Up to 90% diary related diseases are due to

pathogenic bacteria from milk

Dairy industry in most African

countries is dominated by

unpasteurized milk and informal

markets

PCR is a technique which utilizes

primer sets to detect presence of

pathogens in a sample

Risks of milk safety hazards in informal market

are high and unknown in Tanzania

Previous studies have been on the specific risks

associated with pathogenic microbes along the

milk chain (e.g. Swai and Schoonman 2011;

Kaiza et al. 2011)

PCR detection of milk bacterial contaminants is

powerful, gives reliable information on

pathogens in milk

Results of the study will be used to improve food

safety throughout smallholder and informal milk

value chain in Tanzania

Main Objective: To assess milk handling practices, bacterial contamination

and determine selected milk borne zoonotic pathogens along the dairy value chain in Lushoto and Handeni districts of Tanga region

Specific objectives: 1. To assess the possible sources of microbial

contamination of milk from farm to consumer 2. To establish total plate count of bacteria and coliforms

in milk from Lushoto and Handeni districts 3. To establish the prevalence of Escherichia coli

O157:H7 and Brucella abortus in milk using polymerase chain reaction

Study area: Tanga region of the North-eastern

part of Tanzania

Why Tanga?: Well coordinated dairy value chain

◦ Long history of smallholder dairy farming

organized under TDCU

◦ Well developed TFL, one of the biggest

supplier of processed milk in Tanzania

Study design

o A cross–sectional study

Study area

A total of 93 (65 farmers, 28 retailers) interviewed

A total of 166 milk samples from farmers,

vendors, restaurants/kiosks, collection centres

and consumers

Chains identified-Lushoto district

1. Farmer Consumer

2. Farmer Sub Col centre (no chilling ) CC(chilling) Factory Consumer

3. Farmer CC(chilling) Factory Consumer

4. Farmer CC Consumer

5. Farmer CC Restaurant/kiosk Consumer

6. Farmer Selling point Consumer

7. Farmer Restaurant Consumer

8. Farmer/Processor Consumer

Chains identified – Handeni district

1. Farmer Consumer

2. Farmer Vendor Consumer

3. Farmer Vendor Collection Centre Factory consumer

4. Farmer Collection centre consumer

5. Farmer Vendor Restaurant Consumer

Escherichia coli O157:H7 (O157-3 and O157-4)

Brucella abortus (BRU P5 and BRU P8 )

Microbiological isolation, PCR and statistical data analysis (stata IC 11)

Variable Category

No. (%) farmers

respondents

Sources of water Tap 26 (40.0)

Wells 21 (32.3)

Dams and/or streams 19 (29.3)

Milking practices

Cleaning animal shed before

milking 28 (43.1)

Wash hands before milking 46 (70.7)

Wash cow's teats before

milking 41 (63.1)

Wash hands after milking 47 (72.3)

Containers used for milk

storage

wide necked aluminium vessel 2 (03.1)

Wide necked plastic vessel 56 (86.1)

Used water and oil bottles 6 (09.2)

Cooking pan "sufuria" 1 (01.5)

Containers used for

delivery/transportation

wide necked aluminum vessel 0 (0.0)

Wide necked plastick vessel 38 (58.5)

Used water and oil bottles 8 (12.3)

Cooking pan "sufuria" 3 (4.6)

Others e.g traditional pots 16 (24.6)

Means of delivery

On foot 37 (56.9)

By bicycle 9 (13.8)

By motorcycle 3 (4.6)

General practices during milking storage and delivery

Milk marketing channels

Variable Observation

s

Mean

(log10

cfu/ml)

Std. Dev

(log10)

Min Max

Total Plate Count

Farmers 21 5.3 5.4 3.3 5.8

Vendors 5 5.8 5.7 4.6 6.1

Restaurants 7 4.9 4.9 0 5.3

Coliform plate count

Farmers 22 4.8 4.9 2.5 5.5

Vendors 4 4.8 5.1 3.3 5.4

Restaurants 7 3.6 3.9 0 4.3

Total plate counts and coliform plate counts for

milk actors in the chains

Detection of B. abortus by PCR using BRU P5 and BRU P8 primer pairs targeting 16S-23S gene at between 500 to 600 bp. Note that lane M is a molecular weight marker while lanes A, C, D, E, F, G, H, J, K, M, O, P and Q are positive amplicons whereas lane B, I, L and N are negative amplicons. R is a positive control.

42% positive

600 bp500 bp600 bp500 bp

Positive samples for B. abortus in Lushoto and Handeni districts

Source of

milk samples Lushoto (%) Handeni (%)

Both districts (%)

Consumers (2.2) (9.5) (5.7)

Restaurant (4.4) (4.8) (4.6)

Farmers (31.1) (26.2) (28.7)

Vendors - (7.1) (4.6)

Total (37.8) (47.6) (42.5)

Detection of E. coli using O157-3 and O157-4 primer pairs targeting hyla A between at 356 bp. Note that lane M is a molecular weight marker, lane A to K are negative amplicons while lane L is a positive control

M A B C D E F G H I J K L

600 bp500 bp

Poor hygienic practices at milking and selling places contributes to increase in microorganisms

Lack of knowledge on zoonotic diseases and their causes in farmers contributed to poor unhygienic practices in milky activities

The prevalence of B. abortus suggests high contamination rate- relates to findings by Schoonman and Swai (2005)

Veterinary/extension services should be

provided to livestock farmers on proper animal

husbandry and control of diseases.

Responsible authorities must ensure that

existing regulations are instituted and where

possible there should be a mandatory screening

of milk before sales to the public.

Consumer practices such as milk boiling should

be further encouraged.

Further study to relate the findings with human

brucellosis in that area should be carried out.

The Federal Ministry for Economic Cooperation

and Development, Germany through the Safe

Food, Fair Food project

Sokoine University of Agriculture (SUA)

International Livestock Research Institute (ILRI)

AITVM